Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Ethnic variation in genotype frequencies of a p53 intron 7 polymorphism

Ethnic variation in genotype frequencies of a p53 intron 7 polymorphism Abstract Using a PCR–restriction enzyme-based method we found large ethnic variations when a C→T polymorphism in the human p53 gene at position 14181 in intron 7 was studied in Finnish, Polish, Hungarian, Italian, Japanese, Indian and Chinese populations. The largest variations were found between Caucasian and Asian, genotype frequencies varied from C 0.95 and T 0.05 in the Finnish population to C 0.67 and T 0.33 in the Chinese population, the relative risk (RR) for T being 6.5 (95% CI 3.4–12.3, P < 0.001). Variations were also found between Finnish and Italian and between Italian and Chinese (RR for T 2.4, 95% CI 1.2–4.9 and 2.7, 95% CI 1.7–4.2, respectively). Introduction The human tumour suppressor gene p53 has been highly conserved during evolution; only two non-synonymous and three synonymous polymorphisms in the coding regions and eight polymorphisms in the intronic regions have been found (Hernandez-Boussard et al., 1999). Ethnic variations in genotype frequencies of different polymorphisms are common in several genes, such as the human drug metabolizing enzymes cytochrome P-450 (CYP450), glutathione S-transferase (GSTM1) and N-acetyltransferase (NAT2), but there are also reported variations in genotype frequencies of the p53 codon 72 polymorphism (Arg/Pro), that range from 0.79/0.21 in the Mediterranean population to 0.47/0.53 in the African-American population (Jin et al., 1995; To-Figueras et al., 1996). We have previously studied a C→T polymorphism located at position 14181, 72 bp downstream of the 3′-end of exon 7 of the human p53 gene (GenBank accession no. X54156) that is in complete linkage disequilibrium with a T→G polymorphism 20 bp further downstream at position 14201. The study was done in urinary bladder cancer patients and non-cancer controls of Caucasian origin and allele frequencies were C 0.92 and T 0.08 in both groups (Berggren et al., 2000). In this study we have investigated these linked polymorphisms in several ethnic groups to establish variations in genotype frequencies. Materials and methods A total of 485 samples from seven different populations were analysed. DNA was extracted from normal tissue (blood) from 100 Finnish breast cancer patients, 90 healthy Polish, 103 healthy Italian, 26 Indian oral cancer patients, 47 healthy Japanese and 53 (two different groups, n = 37 and 16) healthy Chinese. DNA was extracted from normal tissue surrounding the tumour in 44 Hungarian lung cancer patients and 22 Chinese oesophageal cancer patients. Normal tissue was not available from two of the Chinese oesophageal cancer patients, so tumour tissue was used. We used a PCR–restriction enzyme treatment method previously described in detail (Berggren et al., 2000). Briefly, PCR was performed for 32 cycles in a total volume of 10 μl containing 10 ng genomic DNA and the following primers: sense, 5′-GTT GGC TCT GAC TGT ACC ACC-3′; antisense, 5′-GCC GGA AAT GTG ATG AGA-3′. After PCR samples were treated with the restriction enzyme Eco47I (an isoschizomer of AvaII) that specifically cleaves at 5′-G↓GT/ACC-3′. Thus only samples heterozygous for CT or homozygous for TT at 14181 were cut, leaving two fragments of 180 and 61 bp for the TT homozygous samples and an additional 241 bp fragment from the C allele for CT heterozygous samples. Polymorphisms were easily detected in a 5% PAGE gel stained with ethidium bromide (see Figure 1). All samples that were heterozygous CT at position 14181 were re-amplified by a second PCR and treated with the restriction enzyme Bsp120I (a neoschizomer of ApaI) that specifically cleaves at 5′-G↓GGCCC-3′. Thus only samples that were CC homozygous or CT heterozygous at 14181 were cut, leaving two fragments of 179 and 62 bp. Bsp120I was used to exclude the possibility that the CT heterozygous samples were the result of a partial Eco47I cut of TT homozygous samples. We found no samples that were partially cut with Eco47I, indicating that all detected heterozygous samples were true heterozygotes. Genotype frequencies were calculated from observed genotypes and then compared with the Finnish population as a reference group (lowest allele frequency for T). Expected genotype frequencies were calculated from the Hardy–Weinberg equilibrium (HWE): p2 + 2pq + q2 = 1, where p and q were the observed genotype frequencies (p + q = 1). Observed and expected genotypes were compared with a Pearson χ2 test as follows: Q = (observed CC – expected CC)2/expected CC + (observed CT – expected CT)2/ expected CT + (observed TT – expected TT)2/ expected TT, using χ2 3.84, df 1 [three observations, one estimated parameter (p), –1], α 0.05 (Microsoft Excel 97) (Hartl and Clark, 1997). The Mantel–Haenszel method gave the 95% confidence interval for RR and Fisher's exact test was used to obtain P. Results and discussion Large differences were found between the genotypes C and T at position 14181 of the p53 gene among the different populations (see Table I). Significant differences in genotype frequencies were found between the Finnish population, compared with the Italian, Indian, Japanese and Chinese populations; and also between the Italian, Japanese and Chinese populations, but not between the Italian and Indian populations. Relative risk (RR), confidence intervals (CI) and P values are shown in Table I. Three samples failed to amplify repeatedly (one Finnish, one Polish and one Japanese). These samples were also tested with primers for other genes located on chromosomes other than 17. No amplification was seen for the Japanese sample with any other marker and the two other samples only amplified as short fragments (100–150 bp), whereas longer fragments failed to amplify (>200 bp), suggesting poor quality DNA. These three samples were excluded from the calculations. All genotype frequencies corresponded to the HWE except for two groups of Chinese samples that, for unknown reasons, showed significant deviation from the HWE (see Table I). Also included in Table I is a Swedish population from a previous study of urinary bladder cancer patients and controls. No difference was seen between the two groups and they were treated as one group (Berggren et al., 2000). Previous findings show that there is complete linkage between C at 14181 and T at 14201 and between T at 14181 and G at 14201 (Prosser and Condie, 1991; Berggren et al., 2000). As a test to determine whether this was also true for these samples, tumour tissues from the Chinese oesophageal cancer patients (n = 22) were sequenced using previously reported methods (Berggren et al., 2000). All Chinese oesophageal cancer samples that were sequenced showed complete linkage. Samples that were heterozygous CT (14181) were also heterozygous TG (14201). A representative sequencing picture has been published previously (Berggren et al., 2000). Four heterozygous samples showed loss of heterozygosity (LOH): two of these had lost the C(14181) T(14201) allele and two had lost the T(14181) G(14201) allele. As previously shown, complete linkage disequilibrium was found between the two polymorphisms. Significant linkage disequilibrium is often found for genes and polymorphisms that are tightly linked (i.e. located near each other on the same chromosome). Since there is always an element of chance in the genetic make-up of the two gametes that will form a new generation, chance alone can result in changes in allele frequency, a process known as random genetic drift. Due to random genetic drift each allele faces a risk of extinction in each generation and therefore a single period of small population size, called a population bottleneck, can result in a reduced number of alleles (Hartl and Clark, 1997). The linked polymorphisms we have studied may have resulted from two separate mutations where the original non-mutated allele has disappeared through a population bottleneck. It is also possible that the linked polymorphisms are the result of two simultaneous mutations. Linked polymorphisms can be brought about by recombination, however, since the studied polymorphisms are located only 20 bp apart, recombination between them is extremely rare. Another explanation for linkage disequilibrium is natural selection favouring some genotypes over others, however, because the polymorphisms are located in the intronic sequence this is not likely as an explanation, but cannot be ruled out without functional studies of the protein. In this study of nearly 500 samples we have investigated genotype frequencies for the C→T polymorphism at position 14181, 72 bp downstream of the 3′-end of exon 7 of the human p53 gene, in different ethnic groups. We have shown that genotypes vary 6-fold (proportion with any T, either one, i.e. CT, or two, i.e. TT) for the p53 intron 7 polymorphism between different populations. The variations seem to be largest between groups from different continents, for example Caucasian (Finnish and Swedish) and Asian (Indian, Japanese and Chinese), but there are also differences between Finnish and Italians. Since the genotypes CC, CT and TT (at 14181) do not occur at the same frequency within the different populations, ideally a much larger number of samples than we have analysed should be used to obtain absolute statistical validity for the study. Given the expected frequencies of TT homozygous samples (the smallest frequencies), as a statistical rule of thumb a large enough sample size should be used so that the expected frequency of TT homozygotes (pi) multiplied by the total number of patients (n) should give at least five (n×pi) expected TT homozygous individuals (Blom, 1989). If the polymorphism is associated with a risk of cancer, the genotype frequency in cancer patients will be different from that of the general population. Our data suggests that this is not the case for urinary bladder cancer (Berggren et al., 2000) and we are not aware of any data indicating that this polymorphism elevates cancer risk. The Polish, Italian, Japanese and Chinese (n = 53) samples studied were from healthy individuals and showed a similar variation in polymorphism distribution between populations as did the cancer patients. Thus all available data suggests that our results are valid for the general population. Many reports on ethnic differences in genotype frequencies of polymorphisms have considered the enzymes involved in the activation/inactivation of known carcinogens, such as the CYP450, GSTM1 and NAT2 genes (d'Errico et al., 1996; Smith et al., 1998). These are especially important when they influence mutations in key regulators of cell cycle control, such as the tumour suppressor gene p53 (Ryberg et al., 1994). Individuals who inherit susceptibility alleles of CYP1A1 and GSTM1 have been found to have a high frequency of p53 mutations as a consequence of carcinogen exposure (Kawajiri et al., 1996). The CYP1A1 T5639C polymorphism flanking the 3′-region has only been found in African-American and African populations (Crofts et al., 1993) and the T6235C (flanking the 3′-region) and A4889G (in exon 7 codon 462) polymorphisms are closely, but not completely, linked in Caucasian and Asian (Hayashi et al., 1991; Hirvonen et al., 1992). (Inoue et al., 2000) have studied polymorphisms in CYP1A1 and CYP1B1. They found race-related differences in the occurrence of polymorphisms in both genes and suggested that this may, in part, cause differences in the incidence of lung and breast cancer in Japanese and Caucasian populations. Linked polymorphisms have also been observed in tumour suppressor genes such as CDKN2A (Zhang et al., 1994; Aitken et al., 1999). Linked polymorphisms in exon three of the CDKN2A genes (p16 and p14ARF) have been associated with disease progression in malignant melanoma (Sauroja et al., 2000). Ethnic variations have also been found in the p53 codon 72 Arg/Pro polymorphism (CGC/CCC), where allele frequencies varied from 0.79/0.21 in Mediterranean Caucasians to 0.47/0.53 in African-Americans (Jin et al., 1995; To-Figueras et al., 1996). Allele frequencies in Caucasians (English and Swedish) were 0.78/0.22 and 0.72/0.28 and were all close to 0.60/0.40 in Asians (Chinese and Japanese) (Kawajiri et al., 1993; Birgander et al., 1995; Minaguchi et al., 1998; Rosenthal et al., 1998; Yamashita et al., 1999; Lee et al., 2000). We have detected population variation of a C→T polymorphism in the p53 gene. We did not find any evidence that the studied polymorphism elevates cancer risk. However, studies of polymorphisms may be important because genotype frequencies vary between populations and could possibly explain, in part, the differences in cancer incidence between populations. Intronic polymorphisms probably play a minor role in the field and perhaps the most important polymorphisms are those changing protein function of metabolic enzymes and of cell cycle regulators. Table I. Variation in genotype frequencies for p53 intron 7 C→T polymorphism Population  n  C  T  RR1  CI  P  RR2  CI  P  Q  P  n, number of samples; C = cytosine, T = thymidine at position 14181 of the p53 gene. The Mantel–Haenszel equation was used to calculate RR1,2 (relative risk) and 95% CI (95% confidence interval) and Fisher's exact test was used to obtain the P value. Two comparisons are made; the first uses the Finnish population as the reference group and the second uses the Italian population as the reference group. Observed and expected genotype frequencies were compared using the χ2 test using χ2 3.84, df 1, α 0.05. Test results are represented by Q.  aNormal tissue from cancer patients.  bPreviously published data (Berggren et al. 2000).  cHealthy Chinese.  dChinese with cancer.  eAll Chinese.  Finland  99a  0.949  0.051  1.0  Ref.          2.45  0.12  Swedenb  330  0.923  0.077  1.4  0.7–2.8  0.33        0.56  0.45  Poland  89  0.927  0.073  1.6  0.7–3.4  0.30        0.55  0.46  Hungary  44a  0.909  0.091  1.8  0.7–4.4  0.20        0.44  0.51  Italy  103  0.879  0.121  2.4  1.2–4.9  0.01  1.0  Ref.    0.23  0.63  India  26a  0.769  0.231  4.6  2.1–10.0  <0.001  1.9  1.0–3.5  0.074  0.18  0.67  Japan  46  0.707  0.293  5.8  2.9–11.5  <0.001  2.4  1.5–3.9  <0.001  2.10  0.15  China  37c  0.689  0.311  6.2  3.1–12.3  <0.001  2.6  1.6–4.2  <0.001  3.90  0.05  China  16c  0.656  0.344  6.8  3.1–14.7  <0.001  2.8  1.5–5.2  0.003  0.97  0.32  China  22a,d  0.659  0.341  6.8  3.3–14.0  <0.001  2.8  1.6–4.9  0.001  5.89  0.02  China all  75e  0.673  0.327  6.5  3.4–12.3  <0.001  2.7  1.7–4.2  <0.001  9.93  <0.01  Population  n  C  T  RR1  CI  P  RR2  CI  P  Q  P  n, number of samples; C = cytosine, T = thymidine at position 14181 of the p53 gene. The Mantel–Haenszel equation was used to calculate RR1,2 (relative risk) and 95% CI (95% confidence interval) and Fisher's exact test was used to obtain the P value. Two comparisons are made; the first uses the Finnish population as the reference group and the second uses the Italian population as the reference group. Observed and expected genotype frequencies were compared using the χ2 test using χ2 3.84, df 1, α 0.05. Test results are represented by Q.  aNormal tissue from cancer patients.  bPreviously published data (Berggren et al. 2000).  cHealthy Chinese.  dChinese with cancer.  eAll Chinese.  Finland  99a  0.949  0.051  1.0  Ref.          2.45  0.12  Swedenb  330  0.923  0.077  1.4  0.7–2.8  0.33        0.56  0.45  Poland  89  0.927  0.073  1.6  0.7–3.4  0.30        0.55  0.46  Hungary  44a  0.909  0.091  1.8  0.7–4.4  0.20        0.44  0.51  Italy  103  0.879  0.121  2.4  1.2–4.9  0.01  1.0  Ref.    0.23  0.63  India  26a  0.769  0.231  4.6  2.1–10.0  <0.001  1.9  1.0–3.5  0.074  0.18  0.67  Japan  46  0.707  0.293  5.8  2.9–11.5  <0.001  2.4  1.5–3.9  <0.001  2.10  0.15  China  37c  0.689  0.311  6.2  3.1–12.3  <0.001  2.6  1.6–4.2  <0.001  3.90  0.05  China  16c  0.656  0.344  6.8  3.1–14.7  <0.001  2.8  1.5–5.2  0.003  0.97  0.32  China  22a,d  0.659  0.341  6.8  3.3–14.0  <0.001  2.8  1.6–4.9  0.001  5.89  0.02  China all  75e  0.673  0.327  6.5  3.4–12.3  <0.001  2.7  1.7–4.2  <0.001  9.93  <0.01  View Large Fig. 1. View largeDownload slide 5% PAGE gel showing, from the left: lanes 1–3, treated with the restriction enzyme Eco47I [CC, two intact alleles; CT, one cut allele (T) and one intact allele (C); TT two cut alleles]; lanes 4–6, treated with Bsp120I [CC, two cut alleles; CT, one cut allele (C) and one intact allele (T); TT, two intact alleles]; lane 7, a negative control (–); lane 8, a 50 bp size marker. The arrows point to full-length and cut fragments. Fig. 1. View largeDownload slide 5% PAGE gel showing, from the left: lanes 1–3, treated with the restriction enzyme Eco47I [CC, two intact alleles; CT, one cut allele (T) and one intact allele (C); TT two cut alleles]; lanes 4–6, treated with Bsp120I [CC, two cut alleles; CT, one cut allele (C) and one intact allele (T); TT, two intact alleles]; lane 7, a negative control (–); lane 8, a 50 bp size marker. The arrows point to full-length and cut fragments. 4 To whom correspondence should be addressed at: Department of Biosciences at Novum, Karolinska Institute, 141 57 Huddinge, Sweden. Tel: +46 8 608 92 38; Fax: +46-8–608 15 01; Email: petra.berggren@cnt.ki.se The authors wish to thank Charlotta Granström for help with statistical calculations and Drs Sai-Mei Hou and Kamila Plna for providing samples from healthy Chinese. References Aitken,J., Welch,J., Duffy,D., Milligan,A., Green,A., Martin,N. and Hayward,N. ( 1999) CDKN2A variants in a population-based sample of Queensland families with melanoma. J. Natl Cancer Inst. , 91, 446–452. Google Scholar Berggren,P., Hemminki,K. and Steineck,G. ( 2000) p53 intron 7 polymorphisms in urinary bladder cancer patients and controls. Mutagenesis , 15, 57–60. Google Scholar Birgander,R., Sjalander,A., Rannug,A., Alexandrie,A.K., Sundberg,M.I., Seidegard,J., Tornling,G., Beckman,G. and Beckman,L. ( 1995) P53 polymorphisms and haplotypes in lung cancer. Carcinogenesis , 16, 2233–2236. Google Scholar Blom,G. (1989) Probability and Statistics: Theory and Applications. Springer-Verlag: New York, NY. Google Scholar Crofts,F., Cosma,G.N., Currie,D., Taioli,E., Toniolo,P. and Garte,S.J. (1993) A novel CYP1A1 gene polymorphism in African-Americans. Carcinogenesis , 14, 1729–1731. [Erratum (1993) Carcinogenesis, 14, 2652] Google Scholar d'Errico,A., Taioli,E., Chen,X. and Vineis,P. ( 1996) Genetic metabolic polymorphisms and the risk of cancer: a review of the literature. Biomarkers , 1, 149–173. Google Scholar Hartl,D.L. and Clark,A.G. (1997) Principles of Population Genetics (3 edn). Sinauer Associates, Sunderland, Massachusetts. Google Scholar Hayashi,S., Watanabe,J., Nakachi,K. and Kawajiri,K. ( 1991) Genetic linkage of lung cancer-associated MspI polymorphisms with amino acid replacement in the heme binding region of the human cytochrome P450IA1 gene. J. Biochem. Tokyo , 110, 407–411. Google Scholar Hernandez-Boussard,T., Rodriguez-Tome,P., Montesano,R. and Hainaut,P. ( 1999) IARC p53 mutation database: a relational database to compile and analyze p53 mutations in human tumors and cell lines. Hum. Mutat. , 14, 1–8. Google Scholar Hirvonen,A., Husgafvel-Pursiainen,K., Karjalainen,A., Anttila,S. and Vainio,H. ( 1992) Point-mutational MspI and Ile-Val polymorphisms closely linked in the CYP1A1 gene: lack of association with susceptibility to lung cancer in a Finnish study population. Cancer Epidemiol. Biomarkers Prev. , 1, 485–489. Google Scholar Inoue,K., Asao,T. and Shimada,T. ( 2000) Ethnic-related differences in the frequency distribution of genetic polymorphisms in the CYP1A1 and CYP1B1 genes in Japanese and Caucasian populations. Xenobiotica , 30, 285–295. Google Scholar Jin,X., Wu,X., Roth,J.A., Amos,C.I., King,T.M., Branch,C., Honn,S.E. and Spitz,M.R. ( 1995) Higher lung cancer risk for younger African-Americans with the Pro/Pro p53 genotype. Carcinogenesis , 16, 2205–2208. Google Scholar Kawajiri,K., Nakachi,K., Imai,K., Watanabe,J. and Hayashi,S. ( 1993) Germ line polymorphisms of p53 and CYP1A1 genes involved in human lung cancer. Carcinogenesis , 14, 1085–1089. Google Scholar Kawajiri,K., Eguchi,H., Nakachi,K., Sekiya,T. and Yamamoto,M. ( 1996) Association of CYP1A1 germ line polymorphisms with mutations of the p53 gene in lung cancer. Cancer Res. , 56, 72–76. Google Scholar Lee,J.M., Lee,Y.C., Yang,S.Y., Shi,W.L., Lee,C.J., Luh,S.P., Chen,C.J., Hsieh,C.Y. and Wu,M.T. ( 2000) Genetic polymorphisms of p53 and GSTP1, but not NAT2, are associated with susceptibility to squamous-cell carcinoma of the esophagus. Int. J. Cancer , 89, 458–464. Google Scholar Minaguchi,T., Kanamori,Y., Matsushima,M., Yoshikawa,H., Taketani,Y. and Nakamura,Y. ( 1998) No evidence of correlation between polymorphism at codon 72 of p53 and risk of cervical cancer in Japanese patients with human papillomavirus 16/18 infection. Cancer Res. , 58, 4585–4586. Google Scholar Prosser,J. and Condie,A. ( 1991) Biallelic ApaI polymorphism of the human p53 gene (TP53). Nucleic Acids Res. , 19, 4799. Google Scholar Rosenthal,A.N., Ryan,A., Al-Jehani,R.M., Storey,A., Harwood,C.A. and Jacobs,I.J. ( 1998) p53 codon 72 polymorphism and risk of cervical cancer in UK. Lancet , 352, 871–872. Google Scholar Ryberg,D., Kure,E., Lystad,S., Skaug,V., Stangeland,L., Mercy,I., Borresen,A.L. and Haugen,A. ( 1994) p53 mutations in lung tumors: relationship to putative susceptibility markers for cancer. Cancer Res. , 54, 1551–1555. Google Scholar Sauroja,I., Smeds,J., Vlaykova,T., Kumar,R., Talve,L., Hahka-Kemppinen,M., Punnonen,K., Jansen,C.T., Hemminki,K. and Pyrhonen,S. ( 2000) Analysis of G(1)/S checkpoint regulators in metastatic melanoma. Genes Chromosom. Cancer , 28, 404–414. Google Scholar Smith,G., Stubbins,M.J., Harries,L.W. and Wolf,C.R. ( 1998) Molecular genetics of the human cytochrome P450 monooxygenase superfamily. Xenobiotica , 28, 1129–1165. Google Scholar To-Figueras,J., Gene,M., Gomez-Catalan,J., Galan,C., Firvida,J., Fuentes,M., Rodamilans,M., Huguet,E., Estape,J. and Corbella,J. ( 1996) Glutathione-S-transferase M1 and codon 72 p53 polymorphisms in a northwestern Mediterranean population and their relation to lung cancer susceptibility. Cancer Epidemiol. Biomarkers Prev. , 5, 337–342. Google Scholar Yamashita,T., Yaginuma,Y., Saitoh,Y., Kawai,K., Kurakane,T., Hayashi,H. and Ishikawa,M. ( 1999) Codon 72 polymorphism of p53 as a risk factor for patients with human papillomavirus-associated squamous intraepithelial lesions and invasive cancer of the uterine cervix. Carcinogenesis , 20, 1733–1736. Google Scholar Zhang,S.Y., Klein-Szanto,A.J., Sauter,E.R., Shafarenko,M., Mitsunaga,S., Nobori,T., Carson,D.A., Ridge,J.A. and Goodrow,T.L. ( 1994) Higher frequency of alterations in the p16/CDKN2 gene in squamous cell carcinoma cell lines than in primary tumors of the head and neck. Cancer Res. , 54, 5050–5053. Google Scholar © UK Environmental Mutagen Society/Oxford University Press 2001 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Mutagenesis Oxford University Press

Ethnic variation in genotype frequencies of a p53 intron 7 polymorphism

Ethnic variation in genotype frequencies of a p53 intron 7 polymorphism

Mutagenesis , Volume 16 (6) – Nov 1, 2001

Abstract

Abstract Using a PCR–restriction enzyme-based method we found large ethnic variations when a C→T polymorphism in the human p53 gene at position 14181 in intron 7 was studied in Finnish, Polish, Hungarian, Italian, Japanese, Indian and Chinese populations. The largest variations were found between Caucasian and Asian, genotype frequencies varied from C 0.95 and T 0.05 in the Finnish population to C 0.67 and T 0.33 in the Chinese population, the relative risk (RR) for T being 6.5 (95% CI 3.4–12.3, P < 0.001). Variations were also found between Finnish and Italian and between Italian and Chinese (RR for T 2.4, 95% CI 1.2–4.9 and 2.7, 95% CI 1.7–4.2, respectively). Introduction The human tumour suppressor gene p53 has been highly conserved during evolution; only two non-synonymous and three synonymous polymorphisms in the coding regions and eight polymorphisms in the intronic regions have been found (Hernandez-Boussard et al., 1999). Ethnic variations in genotype frequencies of different polymorphisms are common in several genes, such as the human drug metabolizing enzymes cytochrome P-450 (CYP450), glutathione S-transferase (GSTM1) and N-acetyltransferase (NAT2), but there are also reported variations in genotype frequencies of the p53 codon 72 polymorphism (Arg/Pro), that range from 0.79/0.21 in the Mediterranean population to 0.47/0.53 in the African-American population (Jin et al., 1995; To-Figueras et al., 1996). We have previously studied a C→T polymorphism located at position 14181, 72 bp downstream of the 3′-end of exon 7 of the human p53 gene (GenBank accession no. X54156) that is in complete linkage disequilibrium with a T→G polymorphism 20 bp further downstream at position 14201. The study was done in urinary bladder cancer patients and non-cancer controls of Caucasian origin and allele frequencies were C 0.92 and T 0.08 in both groups (Berggren et al., 2000). In this study we have investigated these linked polymorphisms in several ethnic groups to establish variations in genotype frequencies. Materials and methods A total of 485 samples from seven different populations were analysed. DNA was extracted from normal tissue (blood) from 100 Finnish breast cancer patients, 90 healthy Polish, 103 healthy Italian, 26 Indian oral cancer patients, 47 healthy Japanese and 53 (two different groups, n = 37 and 16) healthy Chinese. DNA was extracted from normal tissue surrounding the tumour in 44 Hungarian lung cancer patients and 22 Chinese oesophageal cancer patients. Normal tissue was not available from two of the Chinese oesophageal cancer patients, so tumour tissue was used. We used a PCR–restriction enzyme treatment method previously described in detail (Berggren et al., 2000). Briefly, PCR was performed for 32 cycles in a total volume of 10 μl containing 10 ng genomic DNA and the following primers: sense, 5′-GTT GGC TCT GAC TGT ACC ACC-3′; antisense, 5′-GCC GGA AAT GTG ATG AGA-3′. After PCR samples were treated with the restriction enzyme Eco47I (an isoschizomer of AvaII) that specifically cleaves at 5′-G↓GT/ACC-3′. Thus only samples heterozygous for CT or homozygous for TT at 14181 were cut, leaving two fragments of 180 and 61 bp for the TT homozygous samples and an additional 241 bp fragment from the C allele for CT heterozygous samples. Polymorphisms were easily detected in a 5% PAGE gel stained with ethidium bromide (see Figure 1). All samples that were heterozygous CT at position 14181 were re-amplified by a second PCR and treated with the restriction enzyme Bsp120I (a neoschizomer of ApaI) that specifically cleaves at 5′-G↓GGCCC-3′. Thus only samples that were CC homozygous or CT heterozygous at 14181 were cut, leaving two fragments of 179 and 62 bp. Bsp120I was used to exclude the possibility that the CT heterozygous samples were the result of a partial Eco47I cut of TT homozygous samples. We found no samples that were partially cut with Eco47I, indicating that all detected heterozygous samples were true heterozygotes. Genotype frequencies were calculated from observed genotypes and then compared with the Finnish population as a reference group (lowest allele frequency for T). Expected genotype frequencies were calculated from the Hardy–Weinberg equilibrium (HWE): p2 + 2pq + q2 = 1, where p and q were the observed genotype frequencies (p + q = 1). Observed and expected genotypes were compared with a Pearson χ2 test as follows: Q = (observed CC – expected CC)2/expected CC + (observed CT – expected CT)2/ expected CT + (observed TT – expected TT)2/ expected TT, using χ2 3.84, df 1 [three observations, one estimated parameter (p), –1], α 0.05 (Microsoft Excel 97) (Hartl and Clark, 1997). The Mantel–Haenszel method gave the 95% confidence interval for RR and Fisher's exact test was used to obtain P. Results and discussion Large differences were found between the genotypes C and T at position 14181 of the p53 gene among the different populations (see Table I). Significant differences in genotype frequencies were found between the Finnish population, compared with the Italian, Indian, Japanese and Chinese populations; and also between the Italian, Japanese and Chinese populations, but not between the Italian and Indian populations. Relative risk (RR), confidence intervals (CI) and P values are shown in Table I. Three samples failed to amplify repeatedly (one Finnish, one Polish and one Japanese). These samples were also tested with primers for other genes located on chromosomes other than 17. No amplification was seen for the Japanese sample with any other marker and the two other samples only amplified as short fragments (100–150 bp), whereas longer fragments failed to amplify (>200 bp), suggesting poor quality DNA. These three samples were excluded from the calculations. All genotype frequencies corresponded to the HWE except for two groups of Chinese samples that, for unknown reasons, showed significant deviation from the HWE (see Table I). Also included in Table I is a Swedish population from a previous study of urinary bladder cancer patients and controls. No difference was seen between the two groups and they were treated as one group (Berggren et al., 2000). Previous findings show that there is complete linkage between C at 14181 and T at 14201 and between T at 14181 and G at 14201 (Prosser and Condie, 1991; Berggren et al., 2000). As a test to determine whether this was also true for these samples, tumour tissues from the Chinese oesophageal cancer patients (n = 22) were sequenced using previously reported methods (Berggren et al., 2000). All Chinese oesophageal cancer samples that were sequenced showed complete linkage. Samples that were heterozygous CT (14181) were also heterozygous TG (14201). A representative sequencing picture has been published previously (Berggren et al., 2000). Four heterozygous samples showed loss of heterozygosity (LOH): two of these had lost the C(14181) T(14201) allele and two had lost the T(14181) G(14201) allele. As previously shown, complete linkage disequilibrium was found between the two polymorphisms. Significant linkage disequilibrium is often found for genes and polymorphisms that are tightly linked (i.e. located near each other on the same chromosome). Since there is always an element of chance in the genetic make-up of the two gametes that will form a new generation, chance alone can result in changes in allele frequency, a process known as random genetic drift. Due to random genetic drift each allele faces a risk of extinction in each generation and therefore a single period of small population size, called a population bottleneck, can result in a reduced number of alleles (Hartl and Clark, 1997). The linked polymorphisms we have studied may have resulted from two separate mutations where the original non-mutated allele has disappeared through a population bottleneck. It is also possible that the linked polymorphisms are the result of two simultaneous mutations. Linked polymorphisms can be brought about by recombination, however, since the studied polymorphisms are located only 20 bp apart, recombination between them is extremely rare. Another explanation for linkage disequilibrium is natural selection favouring some genotypes over others, however, because the polymorphisms are located in the intronic sequence this is not likely as an explanation, but cannot be ruled out without functional studies of the protein. In this study of nearly 500 samples we have investigated genotype frequencies for the C→T polymorphism at position 14181, 72 bp downstream of the 3′-end of exon 7 of the human p53 gene, in different ethnic groups. We have shown that genotypes vary 6-fold (proportion with any T, either one, i.e. CT, or two, i.e. TT) for the p53 intron 7 polymorphism between different populations. The variations seem to be largest between groups from different continents, for example Caucasian (Finnish and Swedish) and Asian (Indian, Japanese and Chinese), but there are also differences between Finnish and Italians. Since the genotypes CC, CT and TT (at 14181) do not occur at the same frequency within the different populations, ideally a much larger number of samples than we have analysed should be used to obtain absolute statistical validity for the study. Given the expected frequencies of TT homozygous samples (the smallest frequencies), as a statistical rule of thumb a large enough sample size should be used so that the expected frequency of TT homozygotes (pi) multiplied by the total number of patients (n) should give at least five (n×pi) expected TT homozygous individuals (Blom, 1989). If the polymorphism is associated with a risk of cancer, the genotype frequency in cancer patients will be different from that of the general population. Our data suggests that this is not the case for urinary bladder cancer (Berggren et al., 2000) and we are not aware of any data indicating that this polymorphism elevates cancer risk. The Polish, Italian, Japanese and Chinese (n = 53) samples studied were from healthy individuals and showed a similar variation in polymorphism distribution between populations as did the cancer patients. Thus all available data suggests that our results are valid for the general population. Many reports on ethnic differences in genotype frequencies of polymorphisms have considered the enzymes involved in the activation/inactivation of known carcinogens, such as the CYP450, GSTM1 and NAT2 genes (d'Errico et al., 1996; Smith et al., 1998). These are especially important when they influence mutations in key regulators of cell cycle control, such as the tumour suppressor gene p53 (Ryberg et al., 1994). Individuals who inherit susceptibility alleles of CYP1A1 and GSTM1 have been found to have a high frequency of p53 mutations as a consequence of carcinogen exposure (Kawajiri et al., 1996). The CYP1A1 T5639C polymorphism flanking the 3′-region has only been found in African-American and African populations (Crofts et al., 1993) and the T6235C (flanking the 3′-region) and A4889G (in exon 7 codon 462) polymorphisms are closely, but not completely, linked in Caucasian and Asian (Hayashi et al., 1991; Hirvonen et al., 1992). (Inoue et al., 2000) have studied polymorphisms in CYP1A1 and CYP1B1. They found race-related differences in the occurrence of polymorphisms in both genes and suggested that this may, in part, cause differences in the incidence of lung and breast cancer in Japanese and Caucasian populations. Linked polymorphisms have also been observed in tumour suppressor genes such as CDKN2A (Zhang et al., 1994; Aitken et al., 1999). Linked polymorphisms in exon three of the CDKN2A genes (p16 and p14ARF) have been associated with disease progression in malignant melanoma (Sauroja et al., 2000). Ethnic variations have also been found in the p53 codon 72 Arg/Pro polymorphism (CGC/CCC), where allele frequencies varied from 0.79/0.21 in Mediterranean Caucasians to 0.47/0.53 in African-Americans (Jin et al., 1995; To-Figueras et al., 1996). Allele frequencies in Caucasians (English and Swedish) were 0.78/0.22 and 0.72/0.28 and were all close to 0.60/0.40 in Asians (Chinese and Japanese) (Kawajiri et al., 1993; Birgander et al., 1995; Minaguchi et al., 1998; Rosenthal et al., 1998; Yamashita et al., 1999; Lee et al., 2000). We have detected population variation of a C→T polymorphism in the p53 gene. We did not find any evidence that the studied polymorphism elevates cancer risk. However, studies of polymorphisms may be important because genotype frequencies vary between populations and could possibly explain, in part, the differences in cancer incidence between populations. Intronic polymorphisms probably play a minor role in the field and perhaps the most important polymorphisms are those changing protein function of metabolic enzymes and of cell cycle regulators. Table I. Variation in genotype frequencies for p53 intron 7 C→T polymorphism Population  n  C  T  RR1  CI  P  RR2  CI  P  Q  P  n, number of samples; C = cytosine, T = thymidine at position 14181 of the p53 gene. The Mantel–Haenszel equation was used to calculate RR1,2 (relative risk) and 95% CI (95% confidence interval) and Fisher's exact test was used to obtain the P value. Two comparisons are made; the first uses the Finnish population as the reference group and the second uses the Italian population as the reference group. Observed and expected genotype frequencies were compared using the χ2 test using χ2 3.84, df 1, α 0.05. Test results are represented by Q.  aNormal tissue from cancer patients.  bPreviously published data (Berggren et al. 2000).  cHealthy Chinese.  dChinese with cancer.  eAll Chinese.  Finland  99a  0.949  0.051  1.0  Ref.          2.45  0.12  Swedenb  330  0.923  0.077  1.4  0.7–2.8  0.33        0.56  0.45  Poland  89  0.927  0.073  1.6  0.7–3.4  0.30        0.55  0.46  Hungary  44a  0.909  0.091  1.8  0.7–4.4  0.20        0.44  0.51  Italy  103  0.879  0.121  2.4  1.2–4.9  0.01  1.0  Ref.    0.23  0.63  India  26a  0.769  0.231  4.6  2.1–10.0  <0.001  1.9  1.0–3.5  0.074  0.18  0.67  Japan  46  0.707  0.293  5.8  2.9–11.5  <0.001  2.4  1.5–3.9  <0.001  2.10  0.15  China  37c  0.689  0.311  6.2  3.1–12.3  <0.001  2.6  1.6–4.2  <0.001  3.90  0.05  China  16c  0.656  0.344  6.8  3.1–14.7  <0.001  2.8  1.5–5.2  0.003  0.97  0.32  China  22a,d  0.659  0.341  6.8  3.3–14.0  <0.001  2.8  1.6–4.9  0.001  5.89  0.02  China all  75e  0.673  0.327  6.5  3.4–12.3  <0.001  2.7  1.7–4.2  <0.001  9.93  <0.01  Population  n  C  T  RR1  CI  P  RR2  CI  P  Q  P  n, number of samples; C = cytosine, T = thymidine at position 14181 of the p53 gene. The Mantel–Haenszel equation was used to calculate RR1,2 (relative risk) and 95% CI (95% confidence interval) and Fisher's exact test was used to obtain the P value. Two comparisons are made; the first uses the Finnish population as the reference group and the second uses the Italian population as the reference group. Observed and expected genotype frequencies were compared using the χ2 test using χ2 3.84, df 1, α 0.05. Test results are represented by Q.  aNormal tissue from cancer patients.  bPreviously published data (Berggren et al. 2000).  cHealthy Chinese.  dChinese with cancer.  eAll Chinese.  Finland  99a  0.949  0.051  1.0  Ref.          2.45  0.12  Swedenb  330  0.923  0.077  1.4  0.7–2.8  0.33        0.56  0.45  Poland  89  0.927  0.073  1.6  0.7–3.4  0.30        0.55  0.46  Hungary  44a  0.909  0.091  1.8  0.7–4.4  0.20        0.44  0.51  Italy  103  0.879  0.121  2.4  1.2–4.9  0.01  1.0  Ref.    0.23  0.63  India  26a  0.769  0.231  4.6  2.1–10.0  <0.001  1.9  1.0–3.5  0.074  0.18  0.67  Japan  46  0.707  0.293  5.8  2.9–11.5  <0.001  2.4  1.5–3.9  <0.001  2.10  0.15  China  37c  0.689  0.311  6.2  3.1–12.3  <0.001  2.6  1.6–4.2  <0.001  3.90  0.05  China  16c  0.656  0.344  6.8  3.1–14.7  <0.001  2.8  1.5–5.2  0.003  0.97  0.32  China  22a,d  0.659  0.341  6.8  3.3–14.0  <0.001  2.8  1.6–4.9  0.001  5.89  0.02  China all  75e  0.673  0.327  6.5  3.4–12.3  <0.001  2.7  1.7–4.2  <0.001  9.93  <0.01  View Large Fig. 1. View largeDownload slide 5% PAGE gel showing, from the left: lanes 1–3, treated with the restriction enzyme Eco47I [CC, two intact alleles; CT, one cut allele (T) and one intact allele (C); TT two cut alleles]; lanes 4–6, treated with Bsp120I [CC, two cut alleles; CT, one cut allele (C) and one intact allele (T); TT, two intact alleles]; lane 7, a negative control (–); lane 8, a 50 bp size marker. The arrows point to full-length and cut fragments. Fig. 1. View largeDownload slide 5% PAGE gel showing, from the left: lanes 1–3, treated with the restriction enzyme Eco47I [CC, two intact alleles; CT, one cut allele (T) and one intact allele (C); TT two cut alleles]; lanes 4–6, treated with Bsp120I [CC, two cut alleles; CT, one cut allele (C) and one intact allele (T); TT, two intact alleles]; lane 7, a negative control (–); lane 8, a 50 bp size marker. The arrows point to full-length and cut fragments. 4 To whom correspondence should be addressed at: Department of Biosciences at Novum, Karolinska Institute, 141 57 Huddinge, Sweden. Tel: +46 8 608 92 38; Fax: +46-8–608 15 01; Email: petra.berggren@cnt.ki.se The authors wish to thank Charlotta Granström for help with statistical calculations and Drs Sai-Mei Hou and Kamila Plna for providing samples from healthy Chinese. References Aitken,J., Welch,J., Duffy,D., Milligan,A., Green,A., Martin,N. and Hayward,N. ( 1999) CDKN2A variants in a population-based sample of Queensland families with melanoma. J. Natl Cancer Inst. , 91, 446–452. Google Scholar Berggren,P., Hemminki,K. and Steineck,G. ( 2000) p53 intron 7 polymorphisms in urinary bladder cancer patients and controls. Mutagenesis , 15, 57–60. Google Scholar Birgander,R., Sjalander,A., Rannug,A., Alexandrie,A.K., Sundberg,M.I., Seidegard,J., Tornling,G., Beckman,G. and Beckman,L. ( 1995) P53 polymorphisms and haplotypes in lung cancer. Carcinogenesis , 16, 2233–2236. Google Scholar Blom,G. (1989) Probability and Statistics: Theory and Applications. Springer-Verlag: New York, NY. Google Scholar Crofts,F., Cosma,G.N., Currie,D., Taioli,E., Toniolo,P. and Garte,S.J. (1993) A novel CYP1A1 gene polymorphism in African-Americans. Carcinogenesis , 14, 1729–1731. [Erratum (1993) Carcinogenesis, 14, 2652] Google Scholar d'Errico,A., Taioli,E., Chen,X. and Vineis,P. ( 1996) Genetic metabolic polymorphisms and the risk of cancer: a review of the literature. Biomarkers , 1, 149–173. Google Scholar Hartl,D.L. and Clark,A.G. (1997) Principles of Population Genetics (3 edn). Sinauer Associates, Sunderland, Massachusetts. Google Scholar Hayashi,S., Watanabe,J., Nakachi,K. and Kawajiri,K. ( 1991) Genetic linkage of lung cancer-associated MspI polymorphisms with amino acid replacement in the heme binding region of the human cytochrome P450IA1 gene. J. Biochem. Tokyo , 110, 407–411. Google Scholar Hernandez-Boussard,T., Rodriguez-Tome,P., Montesano,R. and Hainaut,P. ( 1999) IARC p53 mutation database: a relational database to compile and analyze p53 mutations in human tumors and cell lines. Hum. Mutat. , 14, 1–8. Google Scholar Hirvonen,A., Husgafvel-Pursiainen,K., Karjalainen,A., Anttila,S. and Vainio,H. ( 1992) Point-mutational MspI and Ile-Val polymorphisms closely linked in the CYP1A1 gene: lack of association with susceptibility to lung cancer in a Finnish study population. Cancer Epidemiol. Biomarkers Prev. , 1, 485–489. Google Scholar Inoue,K., Asao,T. and Shimada,T. ( 2000) Ethnic-related differences in the frequency distribution of genetic polymorphisms in the CYP1A1 and CYP1B1 genes in Japanese and Caucasian populations. Xenobiotica , 30, 285–295. Google Scholar Jin,X., Wu,X., Roth,J.A., Amos,C.I., King,T.M., Branch,C., Honn,S.E. and Spitz,M.R. ( 1995) Higher lung cancer risk for younger African-Americans with the Pro/Pro p53 genotype. Carcinogenesis , 16, 2205–2208. Google Scholar Kawajiri,K., Nakachi,K., Imai,K., Watanabe,J. and Hayashi,S. ( 1993) Germ line polymorphisms of p53 and CYP1A1 genes involved in human lung cancer. Carcinogenesis , 14, 1085–1089. Google Scholar Kawajiri,K., Eguchi,H., Nakachi,K., Sekiya,T. and Yamamoto,M. ( 1996) Association of CYP1A1 germ line polymorphisms with mutations of the p53 gene in lung cancer. Cancer Res. , 56, 72–76. Google Scholar Lee,J.M., Lee,Y.C., Yang,S.Y., Shi,W.L., Lee,C.J., Luh,S.P., Chen,C.J., Hsieh,C.Y. and Wu,M.T. ( 2000) Genetic polymorphisms of p53 and GSTP1, but not NAT2, are associated with susceptibility to squamous-cell carcinoma of the esophagus. Int. J. Cancer , 89, 458–464. Google Scholar Minaguchi,T., Kanamori,Y., Matsushima,M., Yoshikawa,H., Taketani,Y. and Nakamura,Y. ( 1998) No evidence of correlation between polymorphism at codon 72 of p53 and risk of cervical cancer in Japanese patients with human papillomavirus 16/18 infection. Cancer Res. , 58, 4585–4586. Google Scholar Prosser,J. and Condie,A. ( 1991) Biallelic ApaI polymorphism of the human p53 gene (TP53). Nucleic Acids Res. , 19, 4799. Google Scholar Rosenthal,A.N., Ryan,A., Al-Jehani,R.M., Storey,A., Harwood,C.A. and Jacobs,I.J. ( 1998) p53 codon 72 polymorphism and risk of cervical cancer in UK. Lancet , 352, 871–872. Google Scholar Ryberg,D., Kure,E., Lystad,S., Skaug,V., Stangeland,L., Mercy,I., Borresen,A.L. and Haugen,A. ( 1994) p53 mutations in lung tumors: relationship to putative susceptibility markers for cancer. Cancer Res. , 54, 1551–1555. Google Scholar Sauroja,I., Smeds,J., Vlaykova,T., Kumar,R., Talve,L., Hahka-Kemppinen,M., Punnonen,K., Jansen,C.T., Hemminki,K. and Pyrhonen,S. ( 2000) Analysis of G(1)/S checkpoint regulators in metastatic melanoma. Genes Chromosom. Cancer , 28, 404–414. Google Scholar Smith,G., Stubbins,M.J., Harries,L.W. and Wolf,C.R. ( 1998) Molecular genetics of the human cytochrome P450 monooxygenase superfamily. Xenobiotica , 28, 1129–1165. Google Scholar To-Figueras,J., Gene,M., Gomez-Catalan,J., Galan,C., Firvida,J., Fuentes,M., Rodamilans,M., Huguet,E., Estape,J. and Corbella,J. ( 1996) Glutathione-S-transferase M1 and codon 72 p53 polymorphisms in a northwestern Mediterranean population and their relation to lung cancer susceptibility. Cancer Epidemiol. Biomarkers Prev. , 5, 337–342. Google Scholar Yamashita,T., Yaginuma,Y., Saitoh,Y., Kawai,K., Kurakane,T., Hayashi,H. and Ishikawa,M. ( 1999) Codon 72 polymorphism of p53 as a risk factor for patients with human papillomavirus-associated squamous intraepithelial lesions and invasive cancer of the uterine cervix. Carcinogenesis , 20, 1733–1736. Google Scholar Zhang,S.Y., Klein-Szanto,A.J., Sauter,E.R., Shafarenko,M., Mitsunaga,S., Nobori,T., Carson,D.A., Ridge,J.A. and Goodrow,T.L. ( 1994) Higher frequency of alterations in the p16/CDKN2 gene in squamous cell carcinoma cell lines than in primary tumors of the head and neck. Cancer Res. , 54, 5050–5053. Google Scholar © UK Environmental Mutagen Society/Oxford University Press 2001

Loading next page...
 
/lp/oxford-university-press/ethnic-variation-in-genotype-frequencies-of-a-p53-intron-7-7jmuCpF6sE

References (25)

Publisher
Oxford University Press
Copyright
© UK Environmental Mutagen Society/Oxford University Press 2001
ISSN
0267-8357
eISSN
1464-3804
DOI
10.1093/mutage/16.6.475
Publisher site
See Article on Publisher Site

Abstract

Abstract Using a PCR–restriction enzyme-based method we found large ethnic variations when a C→T polymorphism in the human p53 gene at position 14181 in intron 7 was studied in Finnish, Polish, Hungarian, Italian, Japanese, Indian and Chinese populations. The largest variations were found between Caucasian and Asian, genotype frequencies varied from C 0.95 and T 0.05 in the Finnish population to C 0.67 and T 0.33 in the Chinese population, the relative risk (RR) for T being 6.5 (95% CI 3.4–12.3, P < 0.001). Variations were also found between Finnish and Italian and between Italian and Chinese (RR for T 2.4, 95% CI 1.2–4.9 and 2.7, 95% CI 1.7–4.2, respectively). Introduction The human tumour suppressor gene p53 has been highly conserved during evolution; only two non-synonymous and three synonymous polymorphisms in the coding regions and eight polymorphisms in the intronic regions have been found (Hernandez-Boussard et al., 1999). Ethnic variations in genotype frequencies of different polymorphisms are common in several genes, such as the human drug metabolizing enzymes cytochrome P-450 (CYP450), glutathione S-transferase (GSTM1) and N-acetyltransferase (NAT2), but there are also reported variations in genotype frequencies of the p53 codon 72 polymorphism (Arg/Pro), that range from 0.79/0.21 in the Mediterranean population to 0.47/0.53 in the African-American population (Jin et al., 1995; To-Figueras et al., 1996). We have previously studied a C→T polymorphism located at position 14181, 72 bp downstream of the 3′-end of exon 7 of the human p53 gene (GenBank accession no. X54156) that is in complete linkage disequilibrium with a T→G polymorphism 20 bp further downstream at position 14201. The study was done in urinary bladder cancer patients and non-cancer controls of Caucasian origin and allele frequencies were C 0.92 and T 0.08 in both groups (Berggren et al., 2000). In this study we have investigated these linked polymorphisms in several ethnic groups to establish variations in genotype frequencies. Materials and methods A total of 485 samples from seven different populations were analysed. DNA was extracted from normal tissue (blood) from 100 Finnish breast cancer patients, 90 healthy Polish, 103 healthy Italian, 26 Indian oral cancer patients, 47 healthy Japanese and 53 (two different groups, n = 37 and 16) healthy Chinese. DNA was extracted from normal tissue surrounding the tumour in 44 Hungarian lung cancer patients and 22 Chinese oesophageal cancer patients. Normal tissue was not available from two of the Chinese oesophageal cancer patients, so tumour tissue was used. We used a PCR–restriction enzyme treatment method previously described in detail (Berggren et al., 2000). Briefly, PCR was performed for 32 cycles in a total volume of 10 μl containing 10 ng genomic DNA and the following primers: sense, 5′-GTT GGC TCT GAC TGT ACC ACC-3′; antisense, 5′-GCC GGA AAT GTG ATG AGA-3′. After PCR samples were treated with the restriction enzyme Eco47I (an isoschizomer of AvaII) that specifically cleaves at 5′-G↓GT/ACC-3′. Thus only samples heterozygous for CT or homozygous for TT at 14181 were cut, leaving two fragments of 180 and 61 bp for the TT homozygous samples and an additional 241 bp fragment from the C allele for CT heterozygous samples. Polymorphisms were easily detected in a 5% PAGE gel stained with ethidium bromide (see Figure 1). All samples that were heterozygous CT at position 14181 were re-amplified by a second PCR and treated with the restriction enzyme Bsp120I (a neoschizomer of ApaI) that specifically cleaves at 5′-G↓GGCCC-3′. Thus only samples that were CC homozygous or CT heterozygous at 14181 were cut, leaving two fragments of 179 and 62 bp. Bsp120I was used to exclude the possibility that the CT heterozygous samples were the result of a partial Eco47I cut of TT homozygous samples. We found no samples that were partially cut with Eco47I, indicating that all detected heterozygous samples were true heterozygotes. Genotype frequencies were calculated from observed genotypes and then compared with the Finnish population as a reference group (lowest allele frequency for T). Expected genotype frequencies were calculated from the Hardy–Weinberg equilibrium (HWE): p2 + 2pq + q2 = 1, where p and q were the observed genotype frequencies (p + q = 1). Observed and expected genotypes were compared with a Pearson χ2 test as follows: Q = (observed CC – expected CC)2/expected CC + (observed CT – expected CT)2/ expected CT + (observed TT – expected TT)2/ expected TT, using χ2 3.84, df 1 [three observations, one estimated parameter (p), –1], α 0.05 (Microsoft Excel 97) (Hartl and Clark, 1997). The Mantel–Haenszel method gave the 95% confidence interval for RR and Fisher's exact test was used to obtain P. Results and discussion Large differences were found between the genotypes C and T at position 14181 of the p53 gene among the different populations (see Table I). Significant differences in genotype frequencies were found between the Finnish population, compared with the Italian, Indian, Japanese and Chinese populations; and also between the Italian, Japanese and Chinese populations, but not between the Italian and Indian populations. Relative risk (RR), confidence intervals (CI) and P values are shown in Table I. Three samples failed to amplify repeatedly (one Finnish, one Polish and one Japanese). These samples were also tested with primers for other genes located on chromosomes other than 17. No amplification was seen for the Japanese sample with any other marker and the two other samples only amplified as short fragments (100–150 bp), whereas longer fragments failed to amplify (>200 bp), suggesting poor quality DNA. These three samples were excluded from the calculations. All genotype frequencies corresponded to the HWE except for two groups of Chinese samples that, for unknown reasons, showed significant deviation from the HWE (see Table I). Also included in Table I is a Swedish population from a previous study of urinary bladder cancer patients and controls. No difference was seen between the two groups and they were treated as one group (Berggren et al., 2000). Previous findings show that there is complete linkage between C at 14181 and T at 14201 and between T at 14181 and G at 14201 (Prosser and Condie, 1991; Berggren et al., 2000). As a test to determine whether this was also true for these samples, tumour tissues from the Chinese oesophageal cancer patients (n = 22) were sequenced using previously reported methods (Berggren et al., 2000). All Chinese oesophageal cancer samples that were sequenced showed complete linkage. Samples that were heterozygous CT (14181) were also heterozygous TG (14201). A representative sequencing picture has been published previously (Berggren et al., 2000). Four heterozygous samples showed loss of heterozygosity (LOH): two of these had lost the C(14181) T(14201) allele and two had lost the T(14181) G(14201) allele. As previously shown, complete linkage disequilibrium was found between the two polymorphisms. Significant linkage disequilibrium is often found for genes and polymorphisms that are tightly linked (i.e. located near each other on the same chromosome). Since there is always an element of chance in the genetic make-up of the two gametes that will form a new generation, chance alone can result in changes in allele frequency, a process known as random genetic drift. Due to random genetic drift each allele faces a risk of extinction in each generation and therefore a single period of small population size, called a population bottleneck, can result in a reduced number of alleles (Hartl and Clark, 1997). The linked polymorphisms we have studied may have resulted from two separate mutations where the original non-mutated allele has disappeared through a population bottleneck. It is also possible that the linked polymorphisms are the result of two simultaneous mutations. Linked polymorphisms can be brought about by recombination, however, since the studied polymorphisms are located only 20 bp apart, recombination between them is extremely rare. Another explanation for linkage disequilibrium is natural selection favouring some genotypes over others, however, because the polymorphisms are located in the intronic sequence this is not likely as an explanation, but cannot be ruled out without functional studies of the protein. In this study of nearly 500 samples we have investigated genotype frequencies for the C→T polymorphism at position 14181, 72 bp downstream of the 3′-end of exon 7 of the human p53 gene, in different ethnic groups. We have shown that genotypes vary 6-fold (proportion with any T, either one, i.e. CT, or two, i.e. TT) for the p53 intron 7 polymorphism between different populations. The variations seem to be largest between groups from different continents, for example Caucasian (Finnish and Swedish) and Asian (Indian, Japanese and Chinese), but there are also differences between Finnish and Italians. Since the genotypes CC, CT and TT (at 14181) do not occur at the same frequency within the different populations, ideally a much larger number of samples than we have analysed should be used to obtain absolute statistical validity for the study. Given the expected frequencies of TT homozygous samples (the smallest frequencies), as a statistical rule of thumb a large enough sample size should be used so that the expected frequency of TT homozygotes (pi) multiplied by the total number of patients (n) should give at least five (n×pi) expected TT homozygous individuals (Blom, 1989). If the polymorphism is associated with a risk of cancer, the genotype frequency in cancer patients will be different from that of the general population. Our data suggests that this is not the case for urinary bladder cancer (Berggren et al., 2000) and we are not aware of any data indicating that this polymorphism elevates cancer risk. The Polish, Italian, Japanese and Chinese (n = 53) samples studied were from healthy individuals and showed a similar variation in polymorphism distribution between populations as did the cancer patients. Thus all available data suggests that our results are valid for the general population. Many reports on ethnic differences in genotype frequencies of polymorphisms have considered the enzymes involved in the activation/inactivation of known carcinogens, such as the CYP450, GSTM1 and NAT2 genes (d'Errico et al., 1996; Smith et al., 1998). These are especially important when they influence mutations in key regulators of cell cycle control, such as the tumour suppressor gene p53 (Ryberg et al., 1994). Individuals who inherit susceptibility alleles of CYP1A1 and GSTM1 have been found to have a high frequency of p53 mutations as a consequence of carcinogen exposure (Kawajiri et al., 1996). The CYP1A1 T5639C polymorphism flanking the 3′-region has only been found in African-American and African populations (Crofts et al., 1993) and the T6235C (flanking the 3′-region) and A4889G (in exon 7 codon 462) polymorphisms are closely, but not completely, linked in Caucasian and Asian (Hayashi et al., 1991; Hirvonen et al., 1992). (Inoue et al., 2000) have studied polymorphisms in CYP1A1 and CYP1B1. They found race-related differences in the occurrence of polymorphisms in both genes and suggested that this may, in part, cause differences in the incidence of lung and breast cancer in Japanese and Caucasian populations. Linked polymorphisms have also been observed in tumour suppressor genes such as CDKN2A (Zhang et al., 1994; Aitken et al., 1999). Linked polymorphisms in exon three of the CDKN2A genes (p16 and p14ARF) have been associated with disease progression in malignant melanoma (Sauroja et al., 2000). Ethnic variations have also been found in the p53 codon 72 Arg/Pro polymorphism (CGC/CCC), where allele frequencies varied from 0.79/0.21 in Mediterranean Caucasians to 0.47/0.53 in African-Americans (Jin et al., 1995; To-Figueras et al., 1996). Allele frequencies in Caucasians (English and Swedish) were 0.78/0.22 and 0.72/0.28 and were all close to 0.60/0.40 in Asians (Chinese and Japanese) (Kawajiri et al., 1993; Birgander et al., 1995; Minaguchi et al., 1998; Rosenthal et al., 1998; Yamashita et al., 1999; Lee et al., 2000). We have detected population variation of a C→T polymorphism in the p53 gene. We did not find any evidence that the studied polymorphism elevates cancer risk. However, studies of polymorphisms may be important because genotype frequencies vary between populations and could possibly explain, in part, the differences in cancer incidence between populations. Intronic polymorphisms probably play a minor role in the field and perhaps the most important polymorphisms are those changing protein function of metabolic enzymes and of cell cycle regulators. Table I. Variation in genotype frequencies for p53 intron 7 C→T polymorphism Population  n  C  T  RR1  CI  P  RR2  CI  P  Q  P  n, number of samples; C = cytosine, T = thymidine at position 14181 of the p53 gene. The Mantel–Haenszel equation was used to calculate RR1,2 (relative risk) and 95% CI (95% confidence interval) and Fisher's exact test was used to obtain the P value. Two comparisons are made; the first uses the Finnish population as the reference group and the second uses the Italian population as the reference group. Observed and expected genotype frequencies were compared using the χ2 test using χ2 3.84, df 1, α 0.05. Test results are represented by Q.  aNormal tissue from cancer patients.  bPreviously published data (Berggren et al. 2000).  cHealthy Chinese.  dChinese with cancer.  eAll Chinese.  Finland  99a  0.949  0.051  1.0  Ref.          2.45  0.12  Swedenb  330  0.923  0.077  1.4  0.7–2.8  0.33        0.56  0.45  Poland  89  0.927  0.073  1.6  0.7–3.4  0.30        0.55  0.46  Hungary  44a  0.909  0.091  1.8  0.7–4.4  0.20        0.44  0.51  Italy  103  0.879  0.121  2.4  1.2–4.9  0.01  1.0  Ref.    0.23  0.63  India  26a  0.769  0.231  4.6  2.1–10.0  <0.001  1.9  1.0–3.5  0.074  0.18  0.67  Japan  46  0.707  0.293  5.8  2.9–11.5  <0.001  2.4  1.5–3.9  <0.001  2.10  0.15  China  37c  0.689  0.311  6.2  3.1–12.3  <0.001  2.6  1.6–4.2  <0.001  3.90  0.05  China  16c  0.656  0.344  6.8  3.1–14.7  <0.001  2.8  1.5–5.2  0.003  0.97  0.32  China  22a,d  0.659  0.341  6.8  3.3–14.0  <0.001  2.8  1.6–4.9  0.001  5.89  0.02  China all  75e  0.673  0.327  6.5  3.4–12.3  <0.001  2.7  1.7–4.2  <0.001  9.93  <0.01  Population  n  C  T  RR1  CI  P  RR2  CI  P  Q  P  n, number of samples; C = cytosine, T = thymidine at position 14181 of the p53 gene. The Mantel–Haenszel equation was used to calculate RR1,2 (relative risk) and 95% CI (95% confidence interval) and Fisher's exact test was used to obtain the P value. Two comparisons are made; the first uses the Finnish population as the reference group and the second uses the Italian population as the reference group. Observed and expected genotype frequencies were compared using the χ2 test using χ2 3.84, df 1, α 0.05. Test results are represented by Q.  aNormal tissue from cancer patients.  bPreviously published data (Berggren et al. 2000).  cHealthy Chinese.  dChinese with cancer.  eAll Chinese.  Finland  99a  0.949  0.051  1.0  Ref.          2.45  0.12  Swedenb  330  0.923  0.077  1.4  0.7–2.8  0.33        0.56  0.45  Poland  89  0.927  0.073  1.6  0.7–3.4  0.30        0.55  0.46  Hungary  44a  0.909  0.091  1.8  0.7–4.4  0.20        0.44  0.51  Italy  103  0.879  0.121  2.4  1.2–4.9  0.01  1.0  Ref.    0.23  0.63  India  26a  0.769  0.231  4.6  2.1–10.0  <0.001  1.9  1.0–3.5  0.074  0.18  0.67  Japan  46  0.707  0.293  5.8  2.9–11.5  <0.001  2.4  1.5–3.9  <0.001  2.10  0.15  China  37c  0.689  0.311  6.2  3.1–12.3  <0.001  2.6  1.6–4.2  <0.001  3.90  0.05  China  16c  0.656  0.344  6.8  3.1–14.7  <0.001  2.8  1.5–5.2  0.003  0.97  0.32  China  22a,d  0.659  0.341  6.8  3.3–14.0  <0.001  2.8  1.6–4.9  0.001  5.89  0.02  China all  75e  0.673  0.327  6.5  3.4–12.3  <0.001  2.7  1.7–4.2  <0.001  9.93  <0.01  View Large Fig. 1. View largeDownload slide 5% PAGE gel showing, from the left: lanes 1–3, treated with the restriction enzyme Eco47I [CC, two intact alleles; CT, one cut allele (T) and one intact allele (C); TT two cut alleles]; lanes 4–6, treated with Bsp120I [CC, two cut alleles; CT, one cut allele (C) and one intact allele (T); TT, two intact alleles]; lane 7, a negative control (–); lane 8, a 50 bp size marker. The arrows point to full-length and cut fragments. Fig. 1. View largeDownload slide 5% PAGE gel showing, from the left: lanes 1–3, treated with the restriction enzyme Eco47I [CC, two intact alleles; CT, one cut allele (T) and one intact allele (C); TT two cut alleles]; lanes 4–6, treated with Bsp120I [CC, two cut alleles; CT, one cut allele (C) and one intact allele (T); TT, two intact alleles]; lane 7, a negative control (–); lane 8, a 50 bp size marker. The arrows point to full-length and cut fragments. 4 To whom correspondence should be addressed at: Department of Biosciences at Novum, Karolinska Institute, 141 57 Huddinge, Sweden. Tel: +46 8 608 92 38; Fax: +46-8–608 15 01; Email: petra.berggren@cnt.ki.se The authors wish to thank Charlotta Granström for help with statistical calculations and Drs Sai-Mei Hou and Kamila Plna for providing samples from healthy Chinese. References Aitken,J., Welch,J., Duffy,D., Milligan,A., Green,A., Martin,N. and Hayward,N. ( 1999) CDKN2A variants in a population-based sample of Queensland families with melanoma. J. Natl Cancer Inst. , 91, 446–452. Google Scholar Berggren,P., Hemminki,K. and Steineck,G. ( 2000) p53 intron 7 polymorphisms in urinary bladder cancer patients and controls. Mutagenesis , 15, 57–60. Google Scholar Birgander,R., Sjalander,A., Rannug,A., Alexandrie,A.K., Sundberg,M.I., Seidegard,J., Tornling,G., Beckman,G. and Beckman,L. ( 1995) P53 polymorphisms and haplotypes in lung cancer. Carcinogenesis , 16, 2233–2236. Google Scholar Blom,G. (1989) Probability and Statistics: Theory and Applications. Springer-Verlag: New York, NY. Google Scholar Crofts,F., Cosma,G.N., Currie,D., Taioli,E., Toniolo,P. and Garte,S.J. (1993) A novel CYP1A1 gene polymorphism in African-Americans. Carcinogenesis , 14, 1729–1731. [Erratum (1993) Carcinogenesis, 14, 2652] Google Scholar d'Errico,A., Taioli,E., Chen,X. and Vineis,P. ( 1996) Genetic metabolic polymorphisms and the risk of cancer: a review of the literature. Biomarkers , 1, 149–173. Google Scholar Hartl,D.L. and Clark,A.G. (1997) Principles of Population Genetics (3 edn). Sinauer Associates, Sunderland, Massachusetts. Google Scholar Hayashi,S., Watanabe,J., Nakachi,K. and Kawajiri,K. ( 1991) Genetic linkage of lung cancer-associated MspI polymorphisms with amino acid replacement in the heme binding region of the human cytochrome P450IA1 gene. J. Biochem. Tokyo , 110, 407–411. Google Scholar Hernandez-Boussard,T., Rodriguez-Tome,P., Montesano,R. and Hainaut,P. ( 1999) IARC p53 mutation database: a relational database to compile and analyze p53 mutations in human tumors and cell lines. Hum. Mutat. , 14, 1–8. Google Scholar Hirvonen,A., Husgafvel-Pursiainen,K., Karjalainen,A., Anttila,S. and Vainio,H. ( 1992) Point-mutational MspI and Ile-Val polymorphisms closely linked in the CYP1A1 gene: lack of association with susceptibility to lung cancer in a Finnish study population. Cancer Epidemiol. Biomarkers Prev. , 1, 485–489. Google Scholar Inoue,K., Asao,T. and Shimada,T. ( 2000) Ethnic-related differences in the frequency distribution of genetic polymorphisms in the CYP1A1 and CYP1B1 genes in Japanese and Caucasian populations. Xenobiotica , 30, 285–295. Google Scholar Jin,X., Wu,X., Roth,J.A., Amos,C.I., King,T.M., Branch,C., Honn,S.E. and Spitz,M.R. ( 1995) Higher lung cancer risk for younger African-Americans with the Pro/Pro p53 genotype. Carcinogenesis , 16, 2205–2208. Google Scholar Kawajiri,K., Nakachi,K., Imai,K., Watanabe,J. and Hayashi,S. ( 1993) Germ line polymorphisms of p53 and CYP1A1 genes involved in human lung cancer. Carcinogenesis , 14, 1085–1089. Google Scholar Kawajiri,K., Eguchi,H., Nakachi,K., Sekiya,T. and Yamamoto,M. ( 1996) Association of CYP1A1 germ line polymorphisms with mutations of the p53 gene in lung cancer. Cancer Res. , 56, 72–76. Google Scholar Lee,J.M., Lee,Y.C., Yang,S.Y., Shi,W.L., Lee,C.J., Luh,S.P., Chen,C.J., Hsieh,C.Y. and Wu,M.T. ( 2000) Genetic polymorphisms of p53 and GSTP1, but not NAT2, are associated with susceptibility to squamous-cell carcinoma of the esophagus. Int. J. Cancer , 89, 458–464. Google Scholar Minaguchi,T., Kanamori,Y., Matsushima,M., Yoshikawa,H., Taketani,Y. and Nakamura,Y. ( 1998) No evidence of correlation between polymorphism at codon 72 of p53 and risk of cervical cancer in Japanese patients with human papillomavirus 16/18 infection. Cancer Res. , 58, 4585–4586. Google Scholar Prosser,J. and Condie,A. ( 1991) Biallelic ApaI polymorphism of the human p53 gene (TP53). Nucleic Acids Res. , 19, 4799. Google Scholar Rosenthal,A.N., Ryan,A., Al-Jehani,R.M., Storey,A., Harwood,C.A. and Jacobs,I.J. ( 1998) p53 codon 72 polymorphism and risk of cervical cancer in UK. Lancet , 352, 871–872. Google Scholar Ryberg,D., Kure,E., Lystad,S., Skaug,V., Stangeland,L., Mercy,I., Borresen,A.L. and Haugen,A. ( 1994) p53 mutations in lung tumors: relationship to putative susceptibility markers for cancer. Cancer Res. , 54, 1551–1555. Google Scholar Sauroja,I., Smeds,J., Vlaykova,T., Kumar,R., Talve,L., Hahka-Kemppinen,M., Punnonen,K., Jansen,C.T., Hemminki,K. and Pyrhonen,S. ( 2000) Analysis of G(1)/S checkpoint regulators in metastatic melanoma. Genes Chromosom. Cancer , 28, 404–414. Google Scholar Smith,G., Stubbins,M.J., Harries,L.W. and Wolf,C.R. ( 1998) Molecular genetics of the human cytochrome P450 monooxygenase superfamily. Xenobiotica , 28, 1129–1165. Google Scholar To-Figueras,J., Gene,M., Gomez-Catalan,J., Galan,C., Firvida,J., Fuentes,M., Rodamilans,M., Huguet,E., Estape,J. and Corbella,J. ( 1996) Glutathione-S-transferase M1 and codon 72 p53 polymorphisms in a northwestern Mediterranean population and their relation to lung cancer susceptibility. Cancer Epidemiol. Biomarkers Prev. , 5, 337–342. Google Scholar Yamashita,T., Yaginuma,Y., Saitoh,Y., Kawai,K., Kurakane,T., Hayashi,H. and Ishikawa,M. ( 1999) Codon 72 polymorphism of p53 as a risk factor for patients with human papillomavirus-associated squamous intraepithelial lesions and invasive cancer of the uterine cervix. Carcinogenesis , 20, 1733–1736. Google Scholar Zhang,S.Y., Klein-Szanto,A.J., Sauter,E.R., Shafarenko,M., Mitsunaga,S., Nobori,T., Carson,D.A., Ridge,J.A. and Goodrow,T.L. ( 1994) Higher frequency of alterations in the p16/CDKN2 gene in squamous cell carcinoma cell lines than in primary tumors of the head and neck. Cancer Res. , 54, 5050–5053. Google Scholar © UK Environmental Mutagen Society/Oxford University Press 2001

Journal

MutagenesisOxford University Press

Published: Nov 1, 2001

There are no references for this article.